/[pub]/test/html-webhacc/cc.cgi
Suika

Diff of /test/html-webhacc/cc.cgi

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.1 by wakaba, Wed Jun 27 11:08:03 2007 UTC revision 1.58 by wakaba, Mon Jul 21 09:40:59 2008 UTC
# Line 2  Line 2 
2  use strict;  use strict;
3    
4  use lib qw[/home/httpd/html/www/markup/html/whatpm  use lib qw[/home/httpd/html/www/markup/html/whatpm
5             /home/wakaba/work/manakai/lib             /home/wakaba/work/manakai2/lib];
            /home/wakaba/public_html/-temp/wiki/lib];  
6  use CGI::Carp qw[fatalsToBrowser];  use CGI::Carp qw[fatalsToBrowser];
 use Time::HiRes qw/time/;  
7    
8  use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module    require WebHACC::Input;
9    
10  my $http = SuikaWiki::Input::HTTP->new;  {
11      require Message::CGI::HTTP;
12      my $http = Message::CGI::HTTP->new;
13    
14      require WebHACC::Output;
15      my $out = WebHACC::Output->new;
16      $out->handle (*STDOUT);
17      $out->set_utf8;
18    
19  ## TODO: _charset_    if ($http->get_meta_variable ('PATH_INFO') ne '/') {
20        $out->http_error (404);
 my @mode = split m#/#, scalar $http->meta_variable ('PATH_INFO'), -1;  
 shift @mode if @mode and $mode[0] == '';  
 ## TODO: decode unreserved characters  
   
   my $s = $http->parameter ('s');  
   if (length $s > 1000_000) {  
     print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long";  
21      exit;      exit;
22    }    }
23    my $char_length = length $s;    
24    my %time;    load_text_catalog ('en'); ## TODO: conneg
   my $time1;  
   my $time2;  
25    
26    require Message::DOM::DOMImplementation;    $out->set_flush;
27    my $dom = Message::DOM::DOMImplementation->____new;    $out->http_header;
28  #  $| = 1;    $out->html_header;
29    my $doc;    $out->unset_flush;
   my $el;  
   
 if (@mode == 3 and $mode[0] eq 'html' and  
     ($mode[2] eq 'html' or $mode[2] eq 'test')) {  
   print STDOUT "Content-Type: text/plain; charset=utf-8\n\n";  
30    
31    require Encode;    my $input = get_input_document ($http);
32    require Whatpm::HTML;    $out->input ($input);
33    
34    $time1 = time;    require WebHACC::Result;
35    $s = Encode::decode ('utf-8', $s);    my $result = WebHACC::Result->new;
36    $time2 = time;    $result->output ($out);
37    $time{decode} = $time2 - $time1;    $result->{conforming_min} = 1;
38        $result->{conforming_max} = 1;
39    
40    print STDOUT "#errors\n";    $out->html ('<script src="../cc-script.js"></script>');
41    
42    my $onerror = sub {    check_and_print ($input => $result => $out);
43      my (%opt) = @_;    
44      print STDOUT "$opt{line},$opt{column},$opt{type}\n";    $result->generate_result_section;
   };  
   
   $doc = $dom->create_document;  
   $time1 = time;  
   if (length $mode[1]) {  
     $el = $doc->create_element_ns  
         ('http://www.w3.org/1999/xhtml', [undef, $mode[1]]);  
     Whatpm::HTML->set_inner_html ($el, $s, $onerror);  
   } else {  
     Whatpm::HTML->parse_string ($s => $doc, $onerror);  
   }  
   $time2 = time;  
   $time{parse} = $time2 - $time1;  
45    
46    print "#document\n";    $out->nav_list;
47    
   my $out;  
   if ($mode[2] eq 'html') {  
     $time1 = time;  
     $out = Whatpm::HTML->get_inner_html ($el || $doc);  
     $time2 = time;  
     $time{serialize_html} = $time2 - $time1;  
   } else { # test  
     $time1 = time;  
     $out = test_serialize ($el || $doc);  
     $time2 = time;  
     $time{serialize_test} = $time2 - $time1;  
   }  
   print STDOUT Encode::encode ('utf-8', $$out);  
   print STDOUT "\n";  
 } elsif (@mode == 3 and $mode[0] eq 'xhtml' and  
          ($mode[2] eq 'html' or $mode[2] eq 'test')) {  
   print STDOUT "Content-Type: text/plain; charset=utf-8\n\n";  
   
   require Message::DOM::XMLParserTemp;  
   print STDOUT "#errors\n";  
   
   my $onerror = sub {  
     my $err = shift;  
     print STDOUT $err->location->line_number, ",";  
     print STDOUT $err->location->column_number, ",";  
     print STDOUT $err->text, "\n";  
     return 1;  
   };  
   
   open my $fh, '<', \$s;  
   my $time1 = time;  
   $doc = Message::DOM::XMLParserTemp->parse_byte_stream  
       ($fh => $dom, $onerror, charset => 'utf-8');  
   my $time2 = time;  
   $time{parse_xml} = $time2 - $time1;  
   
   print "#document\n";  
   
   my $out;  
   if ($mode[2] eq 'html') {  
     ## TODO: Use XHTML serializer  
     #$out = Whatpm::HTML->get_inner_html ($doc);  
   } else { # test  
     $time1 = time;  
     $out = test_serialize ($doc);  
     $time2 = time;  
     $time{serialize_test} = $time2 - $time1;  
   }  
   print STDOUT Encode::encode ('utf-8', $$out);  
   print STDOUT "\n";  
 } else {  
   print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n404";  
48    exit;    exit;
49  }  }
50    
51    if ($http->parameter ('dom5')) {  sub check_and_print ($$$) {
52      require Whatpm::ContentChecker;    my ($input, $result, $out) = @_;
53      my $onerror = sub {    my $original_input = $out->input;
54        my %opt = @_;    $out->input ($input);
55        print STDOUT get_node_path ($opt{node}) . ';' . $opt{type} . "\n";  
56      };    $input->generate_info_section ($result);
57      print STDOUT "#domerrors\n";  
58      $time1 = time;    $input->generate_transfer_sections ($result);
59      if ($el) {  
60        Whatpm::ContentChecker->check_element ($el, $onerror);    unless (defined $input->{s}) {
61      } else {      $result->{conforming_min} = 0;
62        Whatpm::ContentChecker->check_document ($doc, $onerror);      return;
63      }
64    
65      my $checker_class = {
66        'text/cache-manifest' => 'WebHACC::Language::CacheManifest',
67        'text/css' => 'WebHACC::Language::CSS',
68        'text/html' => 'WebHACC::Language::HTML',
69        'text/x-webidl' => 'WebHACC::Language::WebIDL',
70    
71        'text/xml' => 'WebHACC::Language::XML',
72        'application/atom+xml' => 'WebHACC::Language::XML',
73        'application/rss+xml' => 'WebHACC::Language::XML',
74        'image/svg+xml' => 'WebHACC::Language::XML',
75        'application/xhtml+xml' => 'WebHACC::Language::XML',
76        'application/xml' => 'WebHACC::Language::XML',
77        ## TODO: Should we make all XML MIME Types fall
78        ## into this category?
79    
80        ## NOTE: This type has different model from normal XML types.
81        'application/rdf+xml' => 'WebHACC::Language::XML',
82      }->{$input->{media_type}} || 'WebHACC::Language::Default';
83    
84      eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@";
85      my $checker = $checker_class->new;
86      $checker->input ($input);
87      $checker->output ($out);
88      $checker->result ($result);
89    
90      ## TODO: A cache manifest MUST be text/cache-manifest
91      ## TODO: WebIDL media type "text/x-webidl"
92    
93      $checker->generate_syntax_error_section;
94      $checker->generate_source_string_section;
95    
96      my @subdoc;
97      $checker->onsubdoc (sub {
98        push @subdoc, shift;
99      });
100    
101      $checker->generate_structure_dump_section;
102      $checker->generate_structure_error_section;
103      $checker->generate_additional_sections;
104    
105      my $id_prefix = 0;
106      for my $_subinput (@subdoc) {
107        my $subinput = WebHACC::Input::Subdocument->new (++$id_prefix);
108        $subinput->{$_} = $_subinput->{$_} for keys %$_subinput;
109        $subinput->{base_uri} = $subinput->{container_node}->base_uri
110            unless defined $subinput->{base_uri};
111        $subinput->{parent_input} = $input;
112    
113        $subinput->start_section ($result);
114        check_and_print ($subinput => $result => $out);
115        $subinput->end_section ($result);
116      }
117    
118      $out->input ($original_input);
119    } # check_and_print
120    
121    
122    {
123      my $Msg = {};
124    
125    sub load_text_catalog ($) {
126    #  my $self = shift;
127      my $lang = shift; # MUST be a canonical lang name
128      open my $file, '<:utf8', "cc-msg.$lang.txt"
129          or die "$0: cc-msg.$lang.txt: $!";
130      while (<$file>) {
131        if (s/^([^;]+);([^;]*);//) {
132          my ($type, $cls, $msg) = ($1, $2, $_);
133          $msg =~ tr/\x0D\x0A//d;
134          $Msg->{$type} = [$cls, $msg];
135        }
136      }
137    } # load_text_catalog
138    
139    sub get_text ($;$$) {
140    #  my $self = shift;
141      my ($type, $level, $node) = @_;
142      $type = $level . ':' . $type if defined $level;
143      $level = 'm' unless defined $level;
144      my @arg;
145      {
146        if (defined $Msg->{$type}) {
147          my $msg = $Msg->{$type}->[1];
148          $msg =~ s{<var>\$([0-9]+)</var>}{
149            defined $arg[$1] ? ($arg[$1]) : '(undef)';
150          }ge;                 ##BUG: ^ must be escaped
151          $msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{
152            UNIVERSAL::can ($node, 'get_attribute_ns')
153                ?  ($node->get_attribute_ns (undef, $1)) : ''
154          }ge; ## BUG: ^ must be escaped
155          $msg =~ s{<var>{\@}</var>}{        ## BUG: v must be escaped
156            UNIVERSAL::can ($node, 'value') ? ($node->value) : ''
157          }ge;
158          $msg =~ s{<var>{local-name}</var>}{
159            UNIVERSAL::can ($node, 'manakai_local_name')
160              ? ($node->manakai_local_name) : ''
161          }ge;  ## BUG: ^ must be escaped
162          $msg =~ s{<var>{element-local-name}</var>}{
163            (UNIVERSAL::can ($node, 'owner_element') and
164             $node->owner_element)
165              ?  ($node->owner_element->manakai_local_name)
166              : '' ## BUG: ^ must be escaped
167          }ge;
168          return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg);
169        } elsif ($type =~ s/:([^:]*)$//) {
170          unshift @arg, $1;
171          redo;
172      }      }
     $time2 = time;  
     $time{check} = $time2 - $time1;  
173    }    }
174      return ($type, 'level-'.$level, ($_[0]));
175                                     ## BUG: ^ must be escaped
176    } # get_text
177    
178    }
179    
180    sub get_input_document ($) {
181      my $http = shift;
182    
183      require Message::DOM::DOMImplementation;
184      my $dom = Message::DOM::DOMImplementation->new;
185    
186      require Encode;
187      my $request_uri = Encode::decode ('utf-8', $http->get_parameter ('uri'));
188      my $r = WebHACC::Input->new;
189      if (defined $request_uri and length $request_uri) {
190        my $uri = $dom->create_uri_reference ($request_uri);
191        unless ({
192                 http => 1,
193                }->{lc $uri->uri_scheme}) {
194          $r = WebHACC::Input::Error->new;
195          $r->{uri} = $request_uri;
196          $r->{request_uri} = $request_uri;
197          $r->{error_status_text} = 'URL scheme not allowed';
198        }
199    
200        require Message::Util::HostPermit;
201        my $host_permit = new Message::Util::HostPermit;
202        $host_permit->add_rule (<<EOH);
203    Allow host=suika port=80
204    Deny host=suika
205    Allow host=suika.fam.cx port=80
206    Deny host=suika.fam.cx
207    Deny host=localhost
208    Deny host=*.localdomain
209    Deny ipv4=0.0.0.0/8
210    Deny ipv4=10.0.0.0/8
211    Deny ipv4=127.0.0.0/8
212    Deny ipv4=169.254.0.0/16
213    Deny ipv4=172.0.0.0/11
214    Deny ipv4=192.0.2.0/24
215    Deny ipv4=192.88.99.0/24
216    Deny ipv4=192.168.0.0/16
217    Deny ipv4=198.18.0.0/15
218    Deny ipv4=224.0.0.0/4
219    Deny ipv4=255.255.255.255/32
220    Deny ipv6=0::0/0
221    Allow host=*
222    EOH
223        unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) {
224          my $r = WebHACC::Input::Error->new;
225          $r->{uri} = $request_uri;
226          $r->{request_uri} = $request_uri;
227          $r->{error_status_text} = 'Connection to the host is forbidden';
228          return $r;
229        }
230    
231    print STDOUT "#log\n";      require LWP::UserAgent;
232    for (qw/decode parse parse_xml serialize_html serialize_xml serialize_test      my $ua = WDCC::LWPUA->new;
233            check/) {      $ua->{wdcc_dom} = $dom;
234      next unless defined $time{$_};      $ua->{wdcc_host_permit} = $host_permit;
235      print STDOUT {      $ua->agent ('Mozilla'); ## TODO: for now.
236        decode => 'bytes->chars',      $ua->parse_head (0);
237        parse => 'html5(chars)->dom5',      $ua->protocols_allowed ([qw/http/]);
238        parse_xml => 'xml1(chars)->dom5',      $ua->max_size (1000_000);
239        serialize_html => 'dom5->html5(char)',      my $req = HTTP::Request->new (GET => $request_uri);
240        serialize_xml => 'dom5->xml1(char)',      $req->header ('Accept-Encoding' => 'identity, *; q=0');
241        serialize_test => 'dom5->test(char)',      my $res = $ua->request ($req);
242        check => 'dom5 check',      ## TODO: 401 sets |is_success| true.
243      }->{$_};      if ($res->is_success or $http->get_parameter ('error-page')) {
244      print STDOUT "\t", $time{$_}, "s\n";        $r->{base_uri} = $res->base; ## NOTE: It does check |Content-Base|, |Content-Location|, and <base>. ## TODO: Use our own code!
245      open my $file, '>>', ".manakai-$_.txt" or die ".manakai-$_.txt: $!";        $r->{uri} = $res->request->uri;
246      print $file $char_length, "\t", $time{$_}, "\n";        $r->{request_uri} = $request_uri;
247    }  
248          ## TODO: More strict parsing...
249  exit;        my $ct = $res->header ('Content-Type');
250          if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) {
251  sub test_serialize ($) {          $r->{charset} = lc $1;
252    my $node = shift;          $r->{charset} =~ tr/\\//d;
253    my $r = '';          $r->{official_charset} = $r->{charset};
   
   my @node = map { [$_, ''] } @{$node->child_nodes};  
   while (@node) {  
     my $child = shift @node;  
     my $nt = $child->[0]->node_type;  
     if ($nt == $child->[0]->ELEMENT_NODE) {  
       $r .= '| ' . $child->[1] . '<' . $child->[0]->tag_name . ">\x0A"; ## ISSUE: case?  
   
       for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value] }  
                     @{$child->[0]->attributes}) {  
         $r .= '| ' . $child->[1] . '  ' . $attr->[0] . '="'; ## ISSUE: case?  
         $r .= $attr->[1] . '"' . "\x0A";  
254        }        }
255          
256        unshift @node,        my $input_charset = $http->get_parameter ('charset');
257          map { [$_, $child->[1] . '  '] } @{$child->[0]->child_nodes};        if (defined $input_charset and length $input_charset) {
258      } elsif ($nt == $child->[0]->TEXT_NODE) {          $r->{charset_overridden}
259        $r .= '| ' . $child->[1] . '"' . $child->[0]->data . '"' . "\x0A";              = (not defined $r->{charset} or $r->{charset} ne $input_charset);
260      } elsif ($nt == $child->[0]->CDATA_SECTION_NODE) {          $r->{charset} = $input_charset;
261        $r .= '| ' . $child->[1] . '<![CDATA[' . $child->[0]->data . "]]>\x0A";        }
262      } elsif ($nt == $child->[0]->COMMENT_NODE) {  
263        $r .= '| ' . $child->[1] . '<!-- ' . $child->[0]->data . " -->\x0A";        ## TODO: Support for HTTP Content-Encoding
264      } elsif ($nt == $child->[0]->DOCUMENT_TYPE_NODE) {  
265        $r .= '| ' . $child->[1] . '<!DOCTYPE ' . $child->[0]->name . ">\x0A";        $r->{s} = ''.$res->content;
266      } elsif ($nt == $child->[0]->PROCESSING_INSTRUCTION_NODE) {  
267        $r .= '| ' . $child->[1] . '<?' . $child->[0]->target . ' ' .        require Whatpm::ContentType;
268            $child->[0]->data . "?>\x0A";        ($r->{official_type}, $r->{media_type})
269              = Whatpm::ContentType->get_sniffed_type
270                  (get_file_head => sub {
271                     return substr $r->{s}, 0, shift;
272                   },
273                   http_content_type_byte => $ct,
274                   has_http_content_encoding =>
275                       defined $res->header ('Content-Encoding'),
276                   supported_image_types => {});
277      } else {      } else {
278        $r .= '| ' . $child->[1] . $child->[0]->node_type . "\x0A"; # error        $r->{uri} = $res->request->uri;
279          $r->{request_uri} = $request_uri;
280          $r->{error_status_text} = $res->status_line;
281      }      }
282    
283        $r->{header_field} = [];
284        $res->scan (sub {
285          push @{$r->{header_field}}, [$_[0], $_[1]];
286        });
287        $r->{header_status_code} = $res->code;
288        $r->{header_status_text} = $res->message;
289      } else {
290        $r->{s} = ''.$http->get_parameter ('s');
291        $r->{uri} = q<thismessage:/>;
292        $r->{request_uri} = q<thismessage:/>;
293        $r->{base_uri} = q<thismessage:/>;
294        $r->{charset} = ''.$http->get_parameter ('_charset_');
295        $r->{charset} =~ s/\s+//g;
296        $r->{charset} = 'utf-8' if $r->{charset} eq '';
297        $r->{official_charset} = $r->{charset};
298        $r->{header_field} = [];
299    
300        require Whatpm::ContentType;
301        ($r->{official_type}, $r->{media_type})
302            = Whatpm::ContentType->get_sniffed_type
303                (get_file_head => sub {
304                   return substr $r->{s}, 0, shift;
305                 },
306                 http_content_type_byte => undef,
307                 has_http_content_encoding => 0,
308                 supported_image_types => {});
309    }    }
     
   return \$r;  
 } # test_serialize  
310    
311  sub get_node_path ($) {    my $input_format = $http->get_parameter ('i');
312    my $node = shift;    if (defined $input_format and length $input_format) {
313    my @r;      $r->{media_type_overridden}
314    while (defined $node) {          = (not defined $r->{media_type} or $input_format ne $r->{media_type});
315      my $rs;      $r->{media_type} = $input_format;
316      if ($node->node_type == 1) {    }
317        $rs = $node->manakai_local_name;    if (defined $r->{s} and not defined $r->{media_type}) {
318        $node = $node->parent_node;      $r->{media_type} = 'text/html';
319      } elsif ($node->node_type == 2) {      $r->{media_type_overridden} = 1;
320        $rs = '@' . $node->manakai_local_name;    }
321        $node = $node->owner_element;  
322      } elsif ($node->node_type == 3) {    if ($r->{media_type} eq 'text/xml') {
323        $rs = '"' . $node->data . '"';      unless (defined $r->{charset}) {
324        $node = $node->parent_node;        $r->{charset} = 'us-ascii';
325      } elsif ($node->node_type == 9) {        $r->{official_charset} = $r->{charset};
326        $rs = '';      } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') {
327        $node = $node->parent_node;        $r->{charset_overridden} = 0;
     } else {  
       $rs = '#' . $node->node_type;  
       $node = $node->parent_node;  
328      }      }
     unshift @r, $rs;  
329    }    }
330    return join '/', @r;  
331  } # get_node_path    if (length $r->{s} > 1000_000) {
332        $r->{error_status_text} = 'Entity-body too large';
333        delete $r->{s};
334        return $r;
335      }
336    
337      $r->{inner_html_element} = $http->get_parameter ('e');
338    
339      return $r;
340    } # get_input_document
341    
342    package WDCC::LWPUA;
343    BEGIN { push our @ISA, 'LWP::UserAgent'; }
344    
345    sub redirect_ok {
346      my $ua = shift;
347      unless ($ua->SUPER::redirect_ok (@_)) {
348        return 0;
349      }
350    
351      my $uris = $_[1]->header ('Location');
352      return 0 unless $uris;
353      my $uri = $ua->{wdcc_dom}->create_uri_reference ($uris);
354      unless ({
355               http => 1,
356              }->{lc $uri->uri_scheme}) {
357        return 0;
358      }
359      unless ($ua->{wdcc_host_permit}->check ($uri->uri_host, $uri->uri_port || 80)) {
360        return 0;
361      }
362      return 1;
363    } # redirect_ok
364    
365  =head1 AUTHOR  =head1 AUTHOR
366    
# Line 229  Wakaba <w@suika.fam.cx>. Line 368  Wakaba <w@suika.fam.cx>.
368    
369  =head1 LICENSE  =head1 LICENSE
370    
371  Copyright 2007 Wakaba <w@suika.fam.cx>  Copyright 2007-2008 Wakaba <w@suika.fam.cx>
372    
373  This library is free software; you can redistribute it  This library is free software; you can redistribute it
374  and/or modify it under the same terms as Perl itself.  and/or modify it under the same terms as Perl itself.
Legend:
Removed from v.1.1  
changed lines
  Added in v.1.58
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24